Consider the touch button located to the (outside) right of the pull handle (to the left on the inside). The 1.4cm x 2.1cm rectangular sensor is positioned on the inside surface of the 5mm-thick polycarbonate plastic. If we assume this condition can be approximated by an idealized parallel-plate capacitor, we can expect the capacitance to be about C ≈ ε_oε_rA/d = 8.85∙10^-12∙5∙(0.014)∙(0.021)/0.005 = 2.17pF when a finger is pressing against the outer surface. This assumes the area of the finger pressing the outer surface is at least as large as the sensor on the inside surface of the polycarbonate.

We also might approximate the absence of any finger on the touch button as a parallel-plate capacitor with a separation distance much larger than the sensor dimensions. Following this assumption, suppose the distance d is 30mm. This can result in a capacitance of C ≈ ε_oε_rA/d = 8.85∙10^-12∙5∙(0.014)∙(0.021)/0.030 = 0.43pF.

How do these two approximate calculations compare to the demo? The plot in Figure 4-19 shows a plot of the demo's FDC1004 output versus time samples with a finger placed over the touch button between the 400th and 500th sample. Before the finger is applied, the FDC1004 reports the sensor capacitance is about 0.5pF to 0.6pF, which somewhat in agrees with our rough calculations. After the finger is placed over the button, the reported capacitance settles out to 2pF to 2.1pF, which is close to values from our approximate calculations above.